Saint George on a Bike (SGoaB) aims to improve the quality and quantity of open metadata associated with European Cultural Heritage (CH) imagery. To achieve that goal SGoaB addresses two challenges: (i) to transcribe insights about culture, symbols and centuries of evolving iconographic traditions in a knowledge representation accessible to machine learning and artificial intelligence, (ii) to expand conventional machine learning approaches, centered on image recognition, with the ability to decipher the complex pictorial language that characterises iconographic symbols and sacred imagery.